Attempts to introduce genes into hematopoietic stem cells (HSCs) have been disappointing In general, using murine retroviral based vectors to transduce CD34+ cells in vitro have resulted in a low level (0 1-1%) of gene marking for a limited period of time Studies in canines demonstrated that LTBMC transduced with retroviral vectors gave higher levels of gene marking in vivo The advantage of this technique is that by using unfractionated marrow there is no preselection; stem cells and retrovirus are able to colocalize on the stroma; the system selects for LTCIC; transduction occurs 3 times over 21 days following feeding so that cells are in cycle, improving transduction efficiency We examined the utility of this technique in rhesus macaques In 2 animals that underwent this autologous transplantation procedure, no conditioning regime was implemented Unfractionated marrow was transduced for 15 - 18 days with a PA317LN based vector Animals were monitored f or "neo" the marker gene in peripheral blood and bone marrow We observed between 1% and 3% gene marking in peripheral blood for approximately 3 months The frequency of neo in bone marrow was generally lower than that in peripheral blood This technique appears encouraging as our data are similar to what has been observed by others, but our protocol is in the absence of any conditioning regimen Recent efforts have focused on attempts to use LTBMC to transduce HSC using a retroviral vector encoding GFP Initial experiments using a retroviral vector provided by R Howley resulted in a failure to maintain LTBMC, which appears to be due to soluble factors provided by the retroviral producer cell line These results suggest that the LTBMC technique may be difficult to generalize to other retroviral vectors Results of these studies may prove useful in efforts to optimize genetic modification of human hematopoietic stem cells